This note outlines changes to the methodology used to breakdown unallocated or unknown products in trade in goods statistics. This new methodology has been applied to data from January 2021 onwards. In addition, changes have been made to the method for allocating partner country for estimated data.
Intra-EU trade data is collected on the Intrastat survey. Traders are liable to complete the Intrastat survey if their exports are greater than €635k, or imports are greater than €500k in a calendar year. For traders who are below these thresholds their trade is estimated using data from their VAT returns. This is referred to as below threshold trade (BTT). This information is only available at a total value level, and there is no information about partner country or product breakdown. This data was, until now, recorded under SITC9 Commodities and transactions not classified elsewhere. In addition, some estimates for above threshold non-respondents were also allocated to SITC9 if insufficient data was available to create a product profile estimate for them.
Since 2022, under European legislation (European Parliament and Council Regulation (EU) 2019/2152), all EU member states must exchange detailed export micro-data with the corresponding member state of import. The goal of this data exchange is to improve data quality and ultimately to enable Member States to reduce the burden on importers by substituting or supplementing collected data with received micro-data. This new data source has enabled us to create detailed product breakdown for our import estimates.
Note that Northern Ireland is included in this data, as trade with Northern Ireland is collected on the Intrastat survey, even though it is outside the EU.
Two different methods are used to create product breakdowns at individual trader level. For exports a product profile is created at NACE level i.e., the economic classification of activity (Nomenclature of Economic Activities) using the data from Intrastat respondents within that NACE group. The estimated data can then be broken down by product using the NACE of the trader to find a corresponding product profile.
For imports, if there is a match at VAT number with the received microdata then this information is used to estimate the products that they are importing. Approximately 85% of the import data can be matched using this method. Although the import microdata is only available for 2022 the same profile was applied to 2021 data also to create a longer time series for users.
For any importers where there is no match on the received import microdata, a NACE product profile is used similar to the export methodology.
Some trade remains unallocated, this can be due to reported CN codes (Combined Nomenclature) which are classified under SITC9, low value transactions which are not broken down further, and non-matching during the estimation process described above.
Table 1 below outlines the changes to the data at SITC1 level before and after the new methodology has been applied. In fact, the data is broken down into individual CN codes (Combined Nomenclature), however for simplicity only the higher-level aggregates are shown here.
The table shows that for both imports and exports the value of SITC9 (Commodities and transactions not classified elsewhere) decreases whereas all the other headings increase as the value of the unknown trade is reallocated. The value of unallocated trade was significantly higher for imports than exports. This is due to the fact that there is a higher percentage share of below threshold trade for imports than for exports.
It should be noted that the total value of imports and exports has also changed by a small amount in 2022 as part of the normal revision cycle.
| Table 1 Summary of product changes at SITC | €millions | |||||
| 2021 | 2022 | |||||
| Revised Value | Original Value | Change | Revised Value | Original Value | Change | |
| Exports | 165,715 | 165,715 | -0 | 208,435 | 208,445 | -10 |
| 0 Food and live animals | 12,171 | 12,129 | 42 | 14,791 | 14,641 | 150 |
| 1 Beverages and tobacco | 1,762 | 1,759 | 3 | 2,059 | 2,046 | 13 |
| 2 Crude materials, inedible, except fuels | 2,082 | 2,069 | 13 | 2,237 | 2,198 | 39 |
| 3 Mineral fuels, lubricants and related materials | 951 | 948 | 3 | 1,758 | 1,750 | 8 |
| 4 Animal and vegetable oils, fats and waxes | 115 | 114 | 1 | 161 | 155 | 6 |
| 5 Chemicals and related products | 102,637 | 102,603 | 34 | 133,841 | 133,698 | 143 |
| 6 Manufactured goods classified chiefly by material | 2,844 | 2,810 | 34 | 3,377 | 3,272 | 105 |
| 7 Machinery and transport equipment | 25,111 | 25,025 | 86 | 28,291 | 28,024 | 267 |
| 8 Miscellaneous manufactured articles | 17,451 | 17,404 | 47 | 21,330 | 21,175 | 155 |
| 9 Commodities and transactions not classified elsewhere | 591 | 854 | -263 | 590 | 1,485 | -895 |
| Imports | 103,788 | 103,801 | -13 | 141,156 | 141,227 | -71 |
| 0 Food and live animals | 7,505 | 7,415 | 90 | 9,499 | 9,296 | 203 |
| 1 Beverages and tobacco | 1,025 | 1,014 | 11 | 1,175 | 1,149 | 26 |
| 2 Crude materials, inedible, except fuels | 1,107 | 1,059 | 48 | 1,218 | 1,123 | 95 |
| 3 Mineral fuels, lubricants and related materials | 6,188 | 6,183 | 5 | 13,084 | 13,074 | 10 |
| 4 Animal and vegetable oils, fats and waxes | 370 | 369 | 1 | 578 | 574 | 4 |
| 5 Chemicals and related products | 25,986 | 25,904 | 82 | 38,507 | 38,262 | 245 |
| 6 Manufactured goods classified chiefly by material | 7,066 | 6,903 | 163 | 9,362 | 8,940 | 422 |
| 7 Machinery and transport equipment | 41,101 | 40,808 | 293 | 52,279 | 51,619 | 660 |
| 8 Miscellaneous manufactured articles | 12,512 | 12,292 | 220 | 15,193 | 14,707 | 486 |
| 9 Commodities and transactions not classified elsewhere | 928 | 1,854 | -926 | 261 | 2,483 | -2,222 |
Orla McCarthy (+353) 1 498 4091
Ciarán Counihan (+353) 1 498 4215
CSO publication, 21 June 2023, 11am
Learn about our data and confidentiality safeguards, and the steps we take to produce statistics that can be trusted by all.